DNA double-strand breaks are dangerous lesions that must be repaired for cells to survive. Recent studies have demonstrated that error-prone break repair processes such as alternative end joining operate when accurate repair pathways are compromised. These inaccurate processes lead to genomic deletions and chromosome translocations that are associated with genetic diseases and human cancers, especially lymphomas and leukemias. The goal of our research is to characterize molecular mechanisms of inaccurate double-strand break repair and to determine how their utilization may lead to disease. To accomplish this, we propose to elucidate how translesion DNA polymerases contribute to inaccurate repair in the model metazoan Drosophila melanogaster. Specifically, we will investigate a newly-discovered role for DNA polymerase theta in alternative end joining and test the hypothesis that polymerase theta uses coordinated helicase and polymerase activities to promote annealing at microhomologous sequences. Transgenic flies with domain-specific mutations in polymerase theta will be tested for their ability to support alternative end-joining repair. In addition, mutant and wild-type variants of the protein will be expressed in insect cells, purified, and tested in helicase and polymerase assays. In companion experiments, we will use in vivo reporter systems to define the roles of translesion polymerases eta, zeta, and Rev1 in homologous recombination repair of DNA double-strand breaks and gaps that result from I-SceI endonuclease expression and transposon excision. We will also compare the roles of replicative polymerases in break repair requiring various amounts of DNA synthesis and determine how their impairment affects the utilization of translesion polymerases in similar repair contexts. To gain insight into the ways in which local sequence context can affect alternative end joining, we will construct plasmids with I-SceI sites embedded in various DNA repeat contexts and assay repair in embryos expressing I- SceI. Finally, we will characterize a deletion-prone repair pathway that operates in the absence of polymerase theta-mediated alternative end joining and test the model that this backup pathway is a major cause of tumorigenesis in actively dividing tissues of adult flies. Together, these experiments will serve to define how translesion polymerases and associated proteins carry out inaccurate rejoining of broken DNA and will provide important insight into how error-prone double-strand break repair can cause genome instability frequently observed in cancer and related diseases.